This study aims to understand the material behaviour of ore inside a coarse ore stockpile by discrete element modelling. This is of importance since the stockpile is the last stage to control the size distribution of ore in the ore handling system before feeding the grinding circuit. An optimal size distribution is required for efficient grinding in an autogenous grinding (AG) mill. To investigate the material behaviour in the stockpile this work focussed on the flow mode inside the stockpile and the size distribution during discharge. The individual particle characteristics are important to apply in the modelling; hence the discrete element method is very well suited for this type of modelling due to its distinct particle character. In this work the discrete element package used is EDEM, which is a discrete element modelling package developed by the commercial company DEM Solutions. In order to study this aim, several measures in the modelling were taken. The size distribution was divided into the three important sizes, as they are determined by the AG mill. Fines are sized 30mm, pebbles are 80mm and coarse particles are 130mm. A calibration of the parameters of these particles was carried out in order to acquire the correct bulk density, angle of repose and static wall friction of the material in the simulations. Furthermore, an effective simulation model was found by the application of two measures. Firstly, the use of a 3D slice instead of a full 3D simulation showed good results. The second measure was to simulate only coarse sized material to represent each of the particles sizes, which also proved to be very well applicable. Based on segregation mechanisms two different methods to fill the stockpile were determined; batch and mixed filling. The influence of these filling methods was simulated for different flow modes in order to review the impact on the discharged size distribution. It was found that the best suitable mill feed was obtained by batch filled stockpiles, especially since core flow was the expected flow mode. Furthermore, it was found that the ratio between input and output rate has a significant impact on the size distribution at the discharge, since this ratio determines whether there is a ‘last-in, first-out’ system. High input rates result in heap formation, which leads to segregation of coarse sized material. Therefore, at a high input rate it is advised to feed finer material. Feeding of coarser material at moments with a low input rate result in a more optimal size discharge, then when this is done at moments with high input rate.